| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Animal Models |
From the Department of Pathology,*
University of
Washington, Seattle; and the Department of Pediatric
Oncology,
Fred Hutchinson Cancer Research
Center, Seattle, Washington
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-1, or the ligand
glial cell line-derived neurotrophic factor results in renal agenesis.
Transgenic embryos that express a constitutively active form of Ret
(RetMEN2B) under the control of the
dopamine-ß-hydroxylase (DßH) promoter develop profound neuroglial
hyperplasia of their sympathetic ganglia and adrenal medullae. Embryos
from two independent
DßH-RetMEN2B-transgenic
lines exhibit renal malformations. In contrast with
ret-/- embryos, renal maldevelopment in
DßH-RetMEN2B-transgenic
embryos results from primary changes in sympathoadrenal organs
extrinsic to the kidney. The ureteric bud invades the metanephric
mesenchyme normally, but subsequent bud branching and
nephrogenesis are retarded, resulting in severe renal
hypoplasia. Ablation of sympathoadrenal precursors restores normal
renal growth in vivo and in vitro. We
postulate that disruption of renal development results because
RetMEN2B derived from the hyperplastic nervous tissue
competes with endogenous renal Ret for gfr-1 or other signaling
components. This hypothesis is supported by the observation that renal
malformations, which do not normally occur in a transgenic line
with low levels of
DßH-RetMEN2B
expression, arise in a gdnf+/- background.
However, renal maldevelopment was not recapitulated in kidneys
that were co-cultured with explanted transgenic ganglia in
vitro. Our observations illustrate a novel pathogenic mechanism
for renal dysgenesis that may explain how putative activating mutations
of the RET gene can produce a phenotype usually
associated with RET deficiency.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ret functions as one component of a more complex receptor system
for members of the glial cell line-derived neurotrophic factor
(gdnf) family.17
Ligands that activate ret include
gdnf, neurturin, artemin, and persephin. The ligands bind to one or
more gdnf family receptors (gfr-1 to -4), which are
glycosyl-phosphatidyl inositol-linked
proteins. Ligand binding to gfr
probably elicits a conformational change in the latter that affects its
interaction with Ret and thereby promotes Ret
activation.18
Formation of the metanephric kidney and enteric nervous system depends
on interaction between gdnf, Ret, and gfr-1.19-21
The
kidney is formed by reciprocal interaction of two tissues, the
metanephric mesenchyme and the ureteric bud. In murine embryos, the
ureteric bud begins to form on embryonic day (E) 10.5 as an outgrowth
of mesonephric duct cells, which express Ret and gfr-1, in response
to gdnf produced by adjacent metanephric mesenchyme. The ureteric bud
invades the metanephric mesenchyme and goes through a series of
Ret-dependent branching events that culminate in formation of the
entire renal collecting system by E17.22
Concomitant with
invasion and branching by the ureteric bud, metanephric mesenchymal
cells respond to bud-derived inductive signals with a complex series of
morphogenetic events that result in glomerulus and proximal nephron
formation.19,23,24
Ablation of Ret, gfr-1, or gdnf
impairs growth and/or branching of the ureteric bud and results in
renal agenesis or severe dysgenesis.20,21,25-28
Ret-/-, gdnf-/-, and
gfr-1-/- mice also lack intestinal ganglion cells,
indicating a function for this signaling pathway in development of a
second organ system.
MEN2B is characterized by medullary carcinoma of the thyroid,
pheochromocytoma, enteric ganglioneuromas, dysmorphic facies, and
skeletal abnormalities. The clinical syndrome is caused by a point
mutation (M918T) in the tyrosine kinase-coding domain of
RET.6,7,9
The receptor derived from the mutant
allele has constitutive activity which may be further amplified by
gdnf/gfr-1 stimulation.29
The MEN2B mutation also
changes the substrate specificity of the activated receptor, but how
this affects downstream cellular events is not fully
understood.9
To model aspects of the MEN2B phenotype, transgenic mice were produced that overexpress RetMEN2B under control of the dopamine-ß-hydroxylase (DßH) promoter.30 The promoter is active in adrenal chromaffin cells, catecholaminergic neurons, and their precursor cells.31,32 Three lines of transgenic mice were created. All developed benign neuroglial tumors in sympathetic ganglia and adrenal medullae consistent with Ret’s presumed role as a proto-oncogene, although malignant transformation never occurred. Unexpectedly, the majority of the transgenic pups from the intermediate- and high-copy lines displayed renal agenesis or renal dysgenesis, reminiscent of ret-/- mice. Insertional mutagenesis was not the underlying mechanism for the renal phenotype, because it occurred in two independent transgenic lines. Because the DßH promoter is not known to be active within developing kidneys, the transgenic phenotype raised the question of whether extra-renal RetMEN2B overexpression can disrupt renal organogenesis. In this paper, we describe renal development in DßH-RetMEN2B embryos in detail and present evidence that malformation of the kidney is secondary to RetMEN2B expression by hyperplastic sympathoadrenal tissues.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Production and initial characterization of DßH-RetMEN2B and DßH-nlacZ transgenic mice are described elsewhere.30,31 The genetic backgrounds of the founder animals for these transgenic lines were hybrid C57Bl/6J x SJL, but the mice were backcrossed to C57Bl/6J animals for more than four generations. We used two lines of DßH-RetMEN2B-transgenic mice, designated the intermediate-copy line (DßH-RetMEN2B, 70 copies) and the low-copy line (DßH-RetMEN2Blow, 4 copies). Mice were hemizygous for transgenes unless otherwise noted. DßH-RetMEN2Blow, gdnf+/- compound heterozygotes were generated by crossing DßH-RetMEN2Blow mice with mice that carry a null mutation in one gdnf allele (kindly provided by Heiner Westphal, National Institutes of Health).26 Similarly DßH-RetMEN2B, Sox10Dom compound heterozygotes were obtained from crosses of the intermediate-copy line with Dominant megacolon (Sox10Dom,33,34 Jackson Laboratory, Bar Harbor, ME) heterozygotes. For prenatal studies, overnight matings were conducted, with the morning of vaginal plug detection defined as E0.5.
Genotype Analysis
Heads from prenatal mice or 1-cm tail segments from postnatal mice were digested overnight at 37°C in 1x SET buffer (1% sodium dodecyl sulfate, 10 mmol/L Tris-HCl, 5 mmol/L ethylenediaminetetraacetic acid (EDTA), pH 8.0) containing 20 mg/ml proteinase K and 1.4 mmol/L NaCl. DNA was extracted with phenol/chloroform, precipitated in ethanol, and dissolved in 10 mmol/L Tris, 1 mmol/L EDTA, pH 8.0. Polymerase chain reaction (PCR) analysis was performed with 100 ng of DNA as described previously.35 For detection of the DßH-RetMEN2B-transgene primers, 5'-GCA TCT TCA CGG CCA CCG TGG TG-3' and 5'-CGT GGA TGC CTT CAA GAT C-3' were used. PCR analyses to genotype gdnf alleles and the Sox10Dom mice were performed as described by Pichel et al26 and Kapur et al,36 respectively.
Histology, Electron Microscopy, and Immunohistochemistry
Embryos were fixed overnight in 10% buffered formalin, dehydrated, embedded in paraffin, and sectioned. The sections (5 µm) were either stained with hematoxylin and eosin (H&E) or analyzed by immunohistochemistry. For ultrastructural studies, the caudal portion of E12.5-stage embryos were transected just rostral to the lower limb bud, fixed in 3% glutaraldehyde–0.1 mol/L sodium cacodylate (pH 7.4) for 2 hours, and postfixed with osmium tetroxide. The samples were dehydrated, embedded in plastic, and thin-sectioned.
For immunohistochemical staining, sections were deparaffinized, microwave-pretreated for 12 minutes while submerged in 10 mmol/L citrate, pH 6.0, incubated with 0.01% pronase for 15 minutes at 37°C, and blocked with 2% bovine serum albumin in phosphate-buffered saline (PBS) for 10 minutes. Incubation with primary antibody, anti-tyrosine hydroxylase (anti-TH; 1:1000, Eugene Tech International, Ridgefield Park, NJ), or anti-peripherin (1:1000, Chemicon International Inc, Temecula, CA) was performed overnight in a moist chamber at 4°C. Bound antibody was visualized with biotinylated anti-rabbit secondary antibody (90 minutes, 1:250, Boehringer Mannheim, Indianapolis, IN), followed by horseradish peroxidase-conjugated streptavidin (NEN Life Science Products, Boston, MA) and 0.05% diaminobenzidine (Sigma Chemical, St. Louis, MO) in PBS. Retroperitoneal organs from E12.5 and E13.5 DßH-lacZ-transgenic embryos were analyzed for transgene expression by whole-mount X-gal staining as described previously.32
The average number of ureteric bud branches per kidney was calculated from histological counts (three sections per embryo). To determine average kidney size in E14.5 embryos, histological sections of wild-type and transgenic embryos were scanned into Adobe Photoshop 4.0. The image histogram function was then used to calculate the single largest cross-sectional renal area from five different levels. The largest renal cross-sectional area was set at 100%, and the comparative sizes of kidneys from littermates were calculated. Student’s t-tests were applied to evaluate the significance in mean values between different groups.
Organ Culture
Metanephric kidneys were dissected from E12.5 embryos in PBS. Because embryos carrying the DßH-RetMEN2B transgene could not be distinguished grossly from wild-type embryos at this gestational age, cranial tissue was genotyped by PCR analysis. Isolated kidneys were transferred to Millipore filters (0.45-µm pore size). The filters were placed on metal grids in the central wells of 60- x 15-mm tissue culture dishes (Becton Dickinson, San Jose, CA) that contained 1 ml of Dulbecco’s modified Eagle’s medium (DMEM, GIBCO BRL, Grand Island, NY), supplemented with 10% fetal calf serum, 100 U/ml penicillin, and 100 µg/ml streptomycin. The organs were cultured at 37°C in 5% CO2, and the medium was changed every third day. For serum-free cultures, equal volumes of DMEM and Ham’s F-12 supplemented with 6.8 nmol/L selenium, 2 nmol/L triiodothyronine, 0.83 µmol/L insulin, 62 nmol/L transferrin, 71 nmol/L prostaglandin E1, 100 U/ml penicillin, and 100 µg/ml streptomycin were used. After 4 days the organs were either fixed in 10% buffered formalin and embedded for histology or fixed in ice-cold methanol for whole-mount immunohistochemistry.
For celiac ganglion/renal co-cultures, metanephric kidney explants were juxtaposed with preaortic neuroglial tissue from E12.5 embryos and incubated in medium that included serum for 4 days. The embryos used in the co-culture experiments also carried the DßH-nlacZ transgene, expression of which served as a marker of sympathetic neurons. After fixation and X-gal staining, digital images of the explants were captured with a video camera and dissection microscope and analyzed with the image histogram function, as described above. Histological sections of the explants were also prepared and stained with hematoxylin and eosin or immunostained for peripherin immunoreactivity.
Whole-Mount Immunohistochemistry
Pan-cytokeratin whole-mount immunohistochemistry was performed as described by Sainio et al.37 Cultured kidneys were harvested in 80% ice-cold methanol, washed in 11% sucrose–1% bovine serum albumin in PBS, and incubated overnight with an anti-pancytokeratin antibody (1:400, Sigma) at 4°C. After three 2-hour washes in 11% sucrose–1% BSA in PBS, kidneys were incubated overnight with a fluorescein isothiocyanate-conjugated anti-mouse secondary antibody (1:400, Santa Cruz Biotechnology, Santa Cruz, CA) at 4°C. After three 2-hour washes in PBS, kidneys were examined with a fluorescence microscope.
Reverse Transcriptase (RT)-PCR
Metanephric kidneys and intervening soft tissue (celiac ganglia) from E12.5 DßH-RetMEN2B-transgenic and wild-type embryos were harvested individually in 150 µl of lysis buffer, and RNA was isolated with the RNeasy Mini Kit (Qiagen, Santa Clarita, CA), according to the manufacturer’s instructions. Contaminating DNA was digested with 10 U of DNase I, followed by cDNA synthesis with the 1st Strand cDNA Synthesis Kit (Boehringer Mannheim, Indianapolis, IN). Fifty nanograms of cDNA were used per PCR reaction. Primers 5'-GGA GGA GAT GTG TGT CAA CTA TGT GC and 3'-GGG CTC TGA GTC TGT CGG CAT GG were used to amplify endogenous DßH cDNA (350 bp), and primers 5'CCT CTC TGA GGT CCA GGA GG-3' and 5'-GGA GGC TTG AAC AGT GGG ACA TG-3' were used to amplify DHFR cDNA (450 bp). PCR products were separated on 1.5% agarose gels and visualized by ethidium bromide staining.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The DßH promoter drives expression of
ß-galactosidase in DßH-nlacZ-transgenic mice,
thereby providing a histochemical marker for nuclei of
DßH-expressing cells.31,32
To
verify that the DßH promoter is not active
within kidneys during early organogenesis, E12.5 and E13.5
DßH-nlacZ embryos were examined with X-gal
staining. lacZ expression was confirmed in sympathetic
neuroblasts of the preaortic ganglia and within chromaffin cells of the
adrenal gland. However, no blue nuclei were within the developing
kidneys (Figure 1A)
. RT-PCR analysis of
mRNAs isolated from kidneys and sympathoadrenal tissues of
E12.5 transgenic embryos confirmed that the same was true for the
DßH-RetMEN2B transgene.
Although endogenous DßH and transgene-derived
RetMEN2B mRNAs were both detected in
preaortic sympathetic tissue, neither of the genes was expressed within
kidneys (Figure 1B)
.
|
DßH-RetMEN2B-transgenic fetuses survive to term with Mendelian frequency (n = 173), but the postnatal survival rate of transgenic pups was very low. Only 13% survived to weaning (21 days). Twenty transgenic mice that survived to adulthood were sacrificed and examined for renal abnormalities, and three of these had unilateral renal agenesis and a normal contralateral kidney. The remaining 17 had normal kidneys bilaterally.
Most transgenic pups died within the first few days after birth. The frequency of renal anomalies was examined in embryos collected before birth (E17.5) because many dead pups were consumed by their mothers. The vast majority (47 of 62) of transgenic pups had abnormal kidneys, easily distinguished from normal. Most had bilateral renal hypoplasia (38 of 62), whereas unilateral (6 of 62) or bilateral agenesis (2 of 62) was rare. In 6 of 10 cases of renal agenesis, the ureter was present.
Kidney Growth Is Reduced in DßH-RetMEN2B-Transgenic Embyros
Kidney development in wild-type and
DßH-RetMEN2B-transgenic
embryos was investigated by histological analysis of serially sectioned
embryos. Kidneys begin to develop at E10.5 to E11, when the ureteric
bud grows toward the metanephric mesenchyme, and by E11 the ureteric
bud reaches the metanephric mesenchyme. We could not distinguish renal
sections of
DßH-RetMEN2B-transgenic
(n = 8) from wild-type (n
= 3) embryos at E11.5 (not shown). By E12.5, when the ureteric bud had
branched several times within wild-type kidneys
(n = 8; Figures 2A and 3
), branching was reduced in most (14
of 20) DßH-RetMEN2B-transgenic
kidneys (Figures 2B and 3)
. In many transgenic embryos, kidneys
appeared normal on one side, but exhibited reduced or no branching of
the ureteric bud on the opposite side. The ureteric bud almost always
invaded the metanephric mesenchyme (38 of 40 cases), in contrast with
ret-/- embryos, in which the ureteric bud fails to invade
the metanephric mesenchyme (18 of 22).16
|
|
. At later embryonic
stages the size difference was even more pronounced (Figure 4, E and F)
. Histological examination of newborn kidneys confirmed that
hypoplastic kidneys had renal cortices that contained glomeruli,
proximal, and distal tubuli and renal medullae with collecting tubules.
Thus, nephrogenesis occurred, but too few complete nephrons were
formed. Dysplastic features (chondrogenesis, fibrosis, and cysts) were
not observed, although occasional dilated tubules were present in
adults.
|
The DßH promoter is active in sympathetic
neuroblasts from E10.5 onwards. In wild-type E12.5 embryos, preaortic
sympathetic neuroblasts form a solid mass at the level of the
developing adrenal glands. This mass is much larger in
DßH-RetMEN2B-transgenic
embryos, and extends down to the kidney level, where transverse
histological sections of wild-type embryos contain very few neuronal
precursors at this stage (Figure 2, C and E)
. Hence, sympathoadrenal
cells that express the RetMEN2B receptor
were located in close proximity to the developing kidneys (Figure 2, D and F)
, although no direct contact with either the ureteric bud or
the metanephric mesenchyme was observed.
Electron microscopic studies of the celiac region and adjacent tissues
were conducted with E12.5 transgenic and nontransgenic embryos. The
preaortic ganglia from normal embryos and corresponding tumors from
transgenic embryos were morphologically similar. Each was composed of
closely packed cells with abundant cytoplasm that contained numerous
dense-core vesicles, which are typically associated with catecholamine
production (Figure 5A)
. Processes from
these cells were found within the ganglion/tumor, but did not project
into the kidney, even in transgenic embryos with obvious renal
anomalies (Figure 5B)
. Aside from a paucity of ureteric bud-derived
tubules, no renal ultrastructural differences were evident between the
two genotypes (nontransgenic not illustrated).
|
. At this stage, anti-peripherin
immunohistochemistry revealed nerve fibers from the celiac tumor, which
projected in some, but not all, embryos into the developing kidneys
(Figure 6, A and B)
. The nerve bundles
were very prominent at P1 (Figure 6, C and D)
and formed thick bundles
in the renal hilum and cortex. However, distal branches were lacking,
because neither small nerves nor individual fibers were observed
intermixed with surrounding renal tissue.
|
Given that the earliest features of abnormal renal development in
DßH-RetMEN2B-transgenic
embryos were apparent by E12.5, but the transgene was not expressed
within E12.5 kidneys, we investigated whether hypoplasia was an
intrinsic renal defect by examining development of explanted kidneys
in vitro. E12.5 kidneys from transgenic embryos and their
nontransgenic littermates were explanted and cultured in
vitro for 4 days. No difference in overall growth between
wild-type (n = 11) and transgenic
(n = 14) kidneys occurred in serum-containing or
defined culture medium. Histology showed similar nephrogenesis in both
wild-type and transgenic explants (Figure 7, A and B)
, and anti-cytokeratin
whole-mount immunohistochemistry showed no difference in ureteric bud
branching (Figure 7, C and D)
. Thus, it seems unlikely that
a defect intrinsic to the kidney accounts for the renal malformations
in DßH-RetMEN2B-transgenic
mice. Instead, an extrinsic influence, possibly the nearby hyperplasia
of RetMEN2B-expressing sympathoadrenal
precursors, might be involved.
|
To investigate whether hyperplasia of sympathoadrenal precursors
was required for transgene-induced renal malformations, we exploited
the Sox10Dom mutation as a tool to
eliminate prevertebral ganglia in vivo, because homozygous
Sox10Dom/Sox10Dom
mice are missing abdominal sympathetic ganglia. The
Sox10Dom allele resulted from a spontaneous
missense mutation of the Sox10 gene and is associated with
absent enteric ganglia and melanocyte deficiencies in heterozygous
mice.33,34,36,38
Embryos homozygous for the
Sox10Dom mutation die prenatally with severe
deficiencies in their sympathetic nervous system. Preaortic sympathetic
ganglia are virtually absent in
Sox10Dom/Sox10Dom
embryos.39
Introduction of the
DßH-RetMEN2B transgene did not
alter any of these phenotypic features, nor did heterozygosity for a
single Sox10Dom allele affect the renal
malformation rate associated with the
DßH-RetMEN2B transgene (Figure 8)
. However, in
DßH-RetMEN2B,
Sox10Dom/Sox10Dom
embryos, preaortic sympathetic ganglia did not form (Figure 9, B and C)
, and renal malformations did
not occur (8/8; Figures 8 and 9
). Thus, when dissociated from
hyperplastic sympathoadrenal cells induced by the
DßH-RetMEN2B transgene, kidney
growth was normalized in vitro and in vivo.
|
|
In
DßH-RetMEN2B-transgenic
embryos, hyperplastic, preaortic sympathetic neurons form in close
proximity to and later innervate developing kidneys. These neurons
presumably express endogenous Ret and gfr-1, overexpress the
RetMEN2B receptor, and possibly have increased
cellular capacity to bind gdnf. Because kidney formation is dependent
on proper gdnf-ret interaction, a possible mechanism by which
RetMEN2B interferes with renal growth could be
disruption of gdnf-Ret signaling in the kidneys. It is possible that
preaortic sympathetic neurons compete with endogenous Ret receptors
on branches of the ureteric bud for signaling components, reducing
the amount of available gdnf for renal morphogenesis. If so, a more
dramatic phenotype might be achieved experimentally by disrupting one
or both gdnf alleles in
DßH-RetMEN2B embryos. However,
because the frequency of renal malformations already is very high in
DßH-RetMEN2B mice (only 13%
survive to weaning), a worsened phenotype would be difficult to assess.
Instead, we conducted a similar experiment using a low-copy line of
DßH-RetMEN2B
(DßH-RetMEN2Blow).
DßH-RetMEN2Blow-transgenic
mice express less transgene, have sympathetic tumors similar to the
intermediate-copy line, but develop normal kidneys.30
We
crossed
DßH-RetMEN2Blow-transgenic
mice with mice that carry a null mutation in the gdnf
gene.26
As originally described, mice heterozygous for the
null allele sometimes display unilateral agenesis or, rarely,
dysgenesis that does not affect pup viability.26
In our
colony, which has a different genetic background, newborns heterozygous
for the gdnf mutation (n = 16) or
hemizygous for the
DßH-RetMEN2Blow all had normal
kidneys (n = 26; Figure 10
). However, 5 of 12 compound
hemizygous pups
(DßH-RetMEN2Blow,
gdnf+/-) had malformed kidneys (Figure 10)
. Three embryos
had bilaterally small kidneys, and two embryos had unilateral renal
agenesis accompanied by a contralateral hypoplastic kidney. Three
of the compound heterozygotes were found dead at postnatal day 1. Thus,
gdnf haploinsufficiency exposed a renal phenotype in
DßH-RetMEN2Blow embryos, which
was previously associated only with higher levels of transgene
expression in other lines.
|
The preceding data suggest that renal malformations in
DßH-RetMEN2B-transgenic mice
are secondary to sympathoadrenal neuroglial hyperplasia and are
influenced by gdnf expression levels. We suspect that overexpression of
Ret by the neuroglial tumors competes with endogenous renal Ret for
signaling components (gdnf or gfr-1). To test this hypothesis, the
effects of transgenic and nontransgenic sympathoadrenal precursors on
the development of explanted kidneys from E12.5 embryos were compared.
After four days in vitro, no significant differences in
growth (Figure 11)
or histological
differentiation (not shown) were evident between kidneys, regardless of
their genotype or proximity to transgenic neuroglial tissue. A uniform
distribution of differentiated nephron units, derived from metanephric
mesenchyme (glomeruli and proximal tubules) and ureteric bud
(collecting tubules), was present throughout the grafts, including
areas in contact with neuroglial tissue. Thus, the co-culture
conditions used in this study failed to reproduce the renal
malformations observed in vivo.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-1-/- Models
The results of this study suggest that renal malformations in
DßH-RetMEN2B transgenic
embryos are secondary to sympathoadrenal hyperplasia and represent a
distinct pathogenic process from that which underlies renal defects in
ret-/-, gfr-1-/-, or gdnf-/-
embryos. The difference is evident in the renal pathology of
DßH-RetMEN2B pups
versus the knockout models. All of the latter are
characterized predominantly by renal agenesis with severe dysgenesis in
only a small fraction of pups. Renal agenesis was rarely observed in
DßH-RetMEN2B pups, in which
variable renal hypoplasia, ranging from severe to none, was the most
common finding. Ontogeny of renal agenesis has been investigated in
ret-/- embryos and usually results from failure of
ureteric bud formation that is evident by E11.25
When
cells of the metanephric mesenchyme do not receive required bud-derived
signals, they die by apoptosis before E13.5, and no kidneys form.
Renal agenesis and/or dysgenesis in gdnf-/- and
gfr-1-/- mice are also caused by primary failure of the
ureteric bud to form or invade metanephric
mesenchyme.20,21,26-28
In contrast, the first anatomical
indication of defective renal growth in
DßH-RetMEN2B embryos appears
significantly later (E12.5), after successful invasion of the
metanephric blastema by ureteric bud. Beginning at this stage, a subset
of embryos exhibits reduced branching of the ureteric bud and organ
hypoplasia without overt dysplasia (cyst formation, fibrosis, or
chondrogenesis). Thus, the
DßH-RetMEN2B transgene rarely
interferes with the initial formation and growth of the ureteric bud,
but frequently affects later branching events.
Extrinsic Expression of RetMEN2B Interferes with Renal Development
The human DßH promoter in the DßH-RetMEN2B transgene was used to regulate expression of many other cDNAs in transgenic mice, and transgene expression has been restricted consistently to catecholaminergic cells.31,32,40,41 The only exception is a subset of mesenchymal cells in the midface, which transiently express reporter genes fused to the promoter.32 Rare (5–24 cell soma per postnatal kidney) intrinsic neurons have been described in the rat kidney. These neurons are always associated with extrinsic nerves that do not enter the kidney until E16, are primarily located in the renal hilum or adjacent to the pelvis, and co-express TH and DßH with other neurotransmitters.42-45 However, we never observed TH immunoreactivity in the developing murine kidney, except in cell processes of older embryos. Furthermore, no endogenous DßH, DßH-nlacZ, or DßH-RetMEN2B transcripts were detected in E12-E14 renal samples by RT-PCR. Therefore, we believe that the effects of DßH-RetMEN2B on renal development are secondary to gene expression in nonrenal tissues. This hypothesis is supported by the fact that transgenic kidneys exhibit growth and differentiation indistinguishable from nontransgenic kidneys, when cultured as isolated organs in vitro.
The earliest phenotypic change in DßH-RetMEN2B embryos is profound hyperplasia of sympathoadrenal precursors, particularly in the celiac region. The proximity of the primitive neural tumor that arises in this region and the fact that ganglion cells in the normal celiac ganglion innervate the kidneys suggested that sympathoadrenal hyperplasia may be the primary lesion responsible for renal malformations. As predicted by this hypothesis, renal development in Sox10Dom/Sox10Dom, DßH-RetMEN2B embryos, which lack celiac ganglia, was not impaired. Although unanticipated molecular or cellular effects associated with the Sox10Dom/Sox10Dom genotype, other than absent sympathoadrenal precursors, might underlie the rescue phenomenon, we believe that the most straightforward interpretation is that the renal malformations are secondary to sympathoadrenal hyperplasia.
Another approach to test this hypothesis was to co-culture E12.5
celiac ganglia from
DßH-RetMEN2B-transgenic and
nontransgenic embryos with E12.5 renal tissue from transgenic or
nontransgenic mice. Experiments of this type failed to demonstrate any
influence of transgenic sympathoadrenal precursors on renal development
in vitro. It seems likely that either critical in
vivo variables (eg, levels of expression of the transgene,
endogenous ret, gdnf, gfr1, or other signaling components) were not
recapitulated in vitro or hyperplastic preaortic ganglia act
indirectly in vivo via a mechanism that could not be
exercised in the co-cultures (see below).
Competition with Endogenous Ret for gfr-1 or Other Signaling
Components Probably Causes Renal Malformations in
DßH-RetMEN2B Embryos
Several mechanisms might be envisioned by which primary overgrowth of the celiac ganglion could interfere with kidney growth. The tumors might exert a mass effect on the kidneys or their vasculature. Direct physical deformation of renal tissue is unlikely because the body of the tumor does not actually contact the kidney, except through nerve processes, which do not enter the kidney until after renal malformation is evident. Compression of the renal arteries or veins is easier to envision because the tumor abuts these vessels near the embryonic midline. However, our histological studies showed no evidence for arterial narrowing, venous congestion, or thrombosis. Furthermore, structures served by other vessels in the same vicinity (eg, spleen, liver, or lower extremities) exhibited no signs of ischemic injury.
Alternatively, products of the tumor might impair renal development by paracrine, endocrine, or neurosecretory pathways. At E12.5, when retarded branching of the ureteric bud is first detectable, abundant dense-cored neurosecretory granules are present in the celiac masses. Presumably, these contain norepinephrine, which is known to reduce renal blood flow and cell replication in neonatal kidneys and may have comparable effects in developing kidneys.46 Renal hypoplasia in human fetuses has been associated with exposure to pharmacological agents that exhibit vasoactive properties comparable with norepinephrine.47-50 However, the renal pathology differs from DßH-RetMEN2B kidneys in that proximal tubules are markedly reduced or absent, collecting ducts are preserved, and cortical cysts are common.
The preceding mechanisms imply that the role of RetMEN2B in the pathogenesis of renal malformations is merely to promote hyperplasia of preaortic sympathoadrenal precursors. However, the existing data suggest that tumor formation alone is not sufficient. Rather, high expression of DßH-RetMEN2B by the tumors appears necessary. In part, this conclusion is based on the fact that equivalent or greater degrees of preaortic sympathoadrenal overgrowth occur in two other transgenic models, which do not exhibit renal defects. One example is the low-copy line of DßH-RetMEN2Blow mice. The latter carry fewer copies of the DßH-RetMEN2B transgene and express significantly less RetMEN2B protein, but they develop tumors that are ontogenically and pathologically identical to those associated with renal anomalies in the higher-copy line. Another example is mice that carry the DßH-rasval12 transgene, which uses the DßH promoter to target expression of an oncogenic ras cDNA.40 DßH-rasval12 mice develop pathologically similar ganglioneuromas, but their masses are larger, on average, than those of DßH-RetMEN2B mice (R. P. Kapur, unpublished observation). Despite comparable tumors, neither the DßH-RetMEN2Blow nor DßH-rasval12 embryos have kidney malformations. Thus, in addition to sympathoadrenal hyperplasia, something associated with high levels of DßH-RetMEN2B expression appears to impair renal development.
We believe that RetMEN2B protein produced by the preaortic neural tumors of DßH-RetMEN2B transgenic embryos competes for components of the Ret-signaling complex with endogenous Ret made by cells that line branches of the ureteric bud. The effect could be mediated by intact RetMEN2B receptor on the surface of postganglionic nerve fibers or by soluble truncated receptor fragments, which might be secreted by these cells due to receptor overexpression or cDNA truncations within the transgene array. This competitive mechanism could account for transgene-induced renal malformations because outgrowth and branching of the ureteric bud are Ret-dependent. Our conclusion is based partly on the fact that kidney defects do not occur in DßH-RetMEN2Blow embryos, even though they develop preaortic tumors similar in size to the two higher-expressing transgenic lines with renal agenesis. Stronger support for this model is the observation that identical renal anomalies occur in the low-copy line, when embryos are haploinsufficient for gdnf. Thus, the quantity of gdnf produced in gdnf+/- embryos is sufficient to ensure normal renal development in nontransgenic mice, but insufficient to mask renal effects of low levels of DßH-RetMEN2B expression.
Based on this model, it is understandable why the earliest
Ret-dependent event in renal development, outgrowth of the ureteric
bud, is not compromised in most
DßH-RetMEN2B-transgenic
embryos, as it is in gdnf-/-, ret-/-, or
gfr-1-/- embryos. Ureteric bud formation and
initial invasion of the metanephric mesenchyme occur when
sympathoadrenal precursors are just colonizing the preaortic region,
well before a sizable tumor has developed and high levels of transgene
expression are present. Other embryonic events mediated by
gdnf/gfr-1/Ret interaction (eg, enteric neurodevelopment) were not
affected in our embryos, presumably because the
DßH promoter is not active in the embryo until
after Ret-dependent events in enteric neurodevelopment have
concluded.51
A recent independent study led to a similar model to explain renal malformations that result from Ret overexpression in the ureteric bud and other embryonic tissues under control of the Hoxb7 promoter.52 The ontogeny and pathology of renal malformations observed in the latter embryos are similar to DßH-RetMEN2B embryos, but, in contrast with our system, renal maldevelopment occurs in vitro when Hoxb7/RET-transgenic kidneys are explanted. The difference between the two transgenic models strengthens our impression that extrarenal DßH-RetMEN2B transgene expression is responsible for the kidney malformations in our mice.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by Grant RO1DK52530 from the National Institutes of Health and by a postdoctoral fellowship (to C. G.) from the Swedish Cancer Foundation and the Swedish Institute.
Accepted for publication August 24, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
or cyclic AMP-dependent protein kinase A. Mol Cell Biol 1993, 13:358-3663-RET receptor. Neuron 1998, 21:1291-1302[Medline]
1 is an essential receptor component for GDNF in the developing nervous system, and kidney. Neuron 1998, 21:53-62[Medline]
1-deficient mice have deficits in the enteric nervous system, and kidneys. Neuron 1998, 21:317-324[Medline]
This article has been cited by other articles:
 |
|
 |
|
  M. A. Skinner, S. Kalyanaraman, S. D. Safford, R. O. Heuckeroth, W. Tourtellotte, D. Goyeau, P. Goodfellow, J. D. Milbrandt, and A. Freemerman A Human Yeast Artificial Chromosome Containing the Multiple Endocrine Neoplasia Type 2B Ret Mutation Does Not Induce Medullary Thyroid Carcinoma but Does Support the Growth of Kidneys and Partially Rescues Enteric Nervous System Development in Ret-Deficient Mice Am. J. Pathol., January 1, 2005; 166(1): 265 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Bolon, S. Jing, F. Asuncion, S. Scully, M. Pisegna, G. Y. Van, Zheng Hu, Yan Bin Yu, H. Min, K. Wild, et al. The Candidate Neuroprotective Agent Artemin Induces Autonomic Neural Dysplasia without Preventing Peripheral Nerve Dysfunction Toxicol Pathol, April 1, 2004; 32(3): 275 - 294. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE |
